The synthesis, properties and uses of carbon materials with helical morphology

Carbon nanostructures have been widely studied due to their unique properties and potential use in various applications. Of interest has been the study of carbonaceous material with helical morphologies, due to their unique chemical, mechanical, electrical and field emission properties. As such it is envisaged that these materials could be excellent candidates for incorporation in numerous nanotechnology applications. However in order to achieve these aspirations, an understanding of the growth mechanisms and synthetic strategies is necessary. Herein we consider historical and current investigations as reported in the literature, and provide a comprehensive outline of growth mechanisms, synthetic strategies and applications related to helical carbon nanomaterials.

The synthesis, properties and uses of carbon materials

with helical morphology

Ahmed Shaikjee a,b, Neil J Coville a,b,*

a

DST/NRF Centre of Excellence in Strong Materials, University of the Witwatersrand, Johannesburg 2050, South Africab

Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, Johannesburg 2050, South Africa

Received 6 April 2011; revised 21 May 2011; accepted 23 May 2011

Available online 3 August 2011

KEYWORDS

Coiled carbon nanotubes;

Coiled carbon nanofibers;

ª 2011 Cairo University Production and hosting by Elsevier B.V All rights reserved.

Introduction

Carbon is an amazing element, not just because it is the element

required for all life processes, but also due to the fact that it can

exist in numerous allotropic forms[1] Additionally, by means of

synthetic processes, carbon can be tailored into a myriad ofstructures, particularly those in the nanometre range[2–4]

In 1991, Ijima published his landmark paper which describedthe appearance of carbon filaments with diameters in the range

of nanometres[5,6] These carbon materials would come to beknown as carbon nanotubes (CNTs), and play a fundamentalrole in leading scientific and industrial research endeavours innanotechnology Indeed within a matter of years CNTs have ta-ken centre stage in the nano-science arena It is no exaggeration

to say that one of the most active fields of research in the area ofnanotechnology currently is the synthesis, characterization andapplication of CNTs[5,7,8] This has naturally led to a renewedinterest in the synthesis of other forms of carbon nanomaterials:graphene, fibers, horns, buds, onions, helices etc.[8–11] It is thisdiversity in the morphology of carbon materials that providesthe flexibility to modify the properties of carbon Thus, the de-sign and production of carbon materials with unusual morphol-ogies is a promising way to exploit the morphology-propertycorrelation of carbon nano-materials

* Corresponding author Tel.: +27 11 7176738; fax: +27 11 7176749.

E-mail address: Neil.Coville@wits.ac.za (N.J Coville).

2090-1232 ª 2011 Cairo University Production and hosting by

Elsevier B.V All rights reserved.

Peer review under responsibility of Cairo University.

doi: 10.1016/j.jare.2011.05.007

Production and hosting by Elsevier

made based upon a helical design and used by humankind from

ancient times (e.g., the Archimedes water screw) to the present

(e.g., support springs for cellular keypads)[16] It is expected

that nano materials with helical morphology should possess

both similar and unique physical and chemical properties to

their macro components Nano helices should thus behave in

a comparable manner to macro materials with similar

morphol-ogy The ability of a macro scale spring to change shape in

re-sponse to an external force (compression, extension, torsion

etc.), and return to its original shape when the force is removed

has made springs an important component in cellular

technol-ogy, time keeping, medical as well as shock absorbing devices

nano sized springs, the bottom up process starting from atomsand molecules is expected to be the preferred procedure to makethe components needed to form helical nano-materials Thegrowth of helical carbonaceous materials from carbon precur-sors via a bottom up approach in the presence of a catalyst isexpected to proceed by equivalent methods used to synthesizestraight fibers and tubes[5,7] The mechanism commonly pro-posed for carbon fiber growth involves adsorption and dissoci-ation of a carbon precursor on the surface of a catalyst particleand dissolution of carbon into the catalyst particle Once thecatalyst particle has been saturated with carbon, the carboncrystallizes out of the metal particle and is extruded to form a

Fig 1 Various types of helical carbon nanomaterials with non-linear morphology

[3,15,19] The diversity of helical materials provides a myriad

of shaped carbons,Fig 1 The use of helical carbons in

tech-nological applications will be dependent on our ability to

con-trol the coil morphology and coil geometry of these materials

This includes control of the coil diameter, pitch and fiber/tube

thickness,Fig 3c The growth of carbon nano-materials can be

controlled by varying temperature, gas environment and the

type of catalyst The alteration of any of these variables will

result in a significant change in the type and amount of helical

carbon nano-materials formed[3] To achieve this control, an

understanding of the growth mechanism and the role played

by the various parameters is needed To date control over

the synthesis of a specific type of helical carbon nano-material

has been met with only limited success

In this review we attempt to provide a summary of the

var-ious synthetic procedures employed, the relevant mechanistic

explanations that have been given to explain helical growth

patterns and the current technological applications associated

with the new generation of helical carbon nano-materials that

have been prepared In so doing we provide a way forward for

introduction of pentagonal rings into graphene (positive vature) while the insertion of heptagonal and/or octagonalrings led to ‘negative’ curvature [21,22] Before long it wasappreciated that a judicious insertion of a series of pentago-nal and heptagonal rings within a hexagonal matrix wouldyield helically coiled carbon nano-materials As such the is-sue of helical growth is then to achieve the correct combina-tion of polygonal rings (5, 6 and 7) that would generate ahelix [22–25]

cur-Structural origin of helices in CNTs

In order to develop a model that can describe the helical nature

of coiled CNTs, carbon in the form of a fullerene or torus mustfirst be considered Dunlap[21,26]showed that the insertion ofpentagon and heptagon rings at the junction of two CNTs canyield what he called a ‘knee structure’ A knee is formed by thepresence of a pentagon on the convex (positive curvature) sideand of a heptagon on the concave (negative curvature) side of

Fig 2 Arrangement of graphene sheets to produce carbon nanotubes and fibers with various morphologies

a graphene plane, Fig 4 The concept of carbon nanotube

‘knees’ proposed by Dunlap was extended by Fonseca et al

[24]who showed that knee segments could be joined together

to form a toroidal structure (containing 520 carbon atoms,

10 knees) Additionally they were also able to show that if

the knees are joined in such a way that consecutive knees are

joined out of plane, a helix or coil will form instead of a torus

Ihara and Itoh[22]showed that structures that included

pen-tagons and heppen-tagons gave a variety of toroidal structures that

were thermodynamically and energetically stable,Fig 5 They

were able to show that toroidal carbon structures could be used

to model helical CNTs It was noted that the type of toroidal

segment used determines the coil pitch, diameter and cycle of thehelix,Fig 5b (C360) andFig 5c (C540) Additionally they con-cluded that the arrangement of heptagons within the carbon ma-trix was instrumental in controlling the coil geometry A study

by Setton and Setton[27] concluded that while toroidal ments could be used to model helical CNTs, they could only

seg-be used to explain single shell helices or at seg-best two shell helices.They suggested that for multi shelled helices, pentagon and hep-tagon pairs would have to be arranged along the helical path, oralternatively other ‘defects’ would need to be considered Mostrecently Liu et al.[28]were able to demonstrate, using atomisticmodels, that by introducing a pair of pentagons and a pair ofFig 3 Schematic illustration: (a) solid coiled fiber, (b) tubular coiled fiber and (c) parameters used to define coil morphology

heptagons into the structure of a single walled CNT that a

curved structure could be obtained The pair of pentagons forms

a cone defect whereas the pair of heptagons results in a saddle

point The incorporation of the pentagons/heptagons creates

strain, which is released when the CNT bends at the defect site

They suggested that by varying the diameter of the nanotube

and/or the length of the basic segment, the coil diameter, coil

pitch and tubular diameter could be varied Biro´ et al.[25]

at-tempted to explain the incorporation of pentagon/heptagon

pairs by considering the possibility that pentagon/heptagon

pairs were not simply defects but were regular building blocks

for the helical CNT structure They proposed that Haeckelite

type sheets, which are characterized by a high number of

penta-gon/heptagon pairs, could be rolled like a graphene sheet to

yield helical CNTs,Fig 6 Furthermore experimental

observa-tions of Haeckelite type structures indicated that they could be

produced by procedures analogous to those used to generate

CNTs Lu et al.[29]proposed that during the initial growth of

helical CNTs, prevailing reaction conditions would result in

the nucleation of a pentagon, which would result in the

forma-tion of a spiral shell around a catalyst particle,Fig 7 [19] From

this core structure, curved or straight segments emerge that pend upon whether there are only hexagons (straight segment)

de-or pentagon/heptagon pairs (curved segments) present As such,geometric parameters (coil pitch, twist angle etc.) are deter-mined by the frequency of pentagon/heptagon pair creation.While these models are useful, they cannot explain how penta-gon/heptagon pairs can be incorporated in such a manner.Fonseca et al.[24]attempted to explain the introduction of

‘knees’ (pentagon/heptagon pairs) by means of steric drance They proposed that if the growth path of a CNTwas blocked, formation of a knee at the catalyst surface wouldcause a bend in the tube before continued growth,Fig 8 Asfurther blockages were encountered further knees would beintroduced, resulting in regular and irregular helically coiledCNTs However this model has been met with limited accep-tance as blockages would have to be systematic (to ensure reg-ular coiling) and adjacent tubes would be expected to interferewith each other’s helicity as they collided during growth.While the concept of pentagon/heptagon pairs has been ac-cepted as the best model to explain helical growth, Ramachan-dran and Sathyamurthy [30] have suggested that rotationaldistortion of carbon fragments, that do not alter the hexagonalmatrix is also capable of yielding helical CNTs They suggestedthat as a CNT grows, the adjacent layers can undergo rota-tional distortion by some small angle from their originalposition This continued distortion of subsequent layers results

hin-Fig 5 (a) Toroidal structure made up of pentagons and heptagons (C360), (b) helical coil made up of toroidal (C360segments), (c) helicalcoil made up of toroidal (C segments)[22]

Fig 4 Knee formed by pentagon/heptagon pair[24]

Fig 6 Haeckelite structure, graphite sheet composed of onal rings, that can be rolled to form helical nanotubes (based onref[25])

polyg-in a coiled nanotube This mechanism elimpolyg-inates the need for

incorporation of pentagon/heptagon pairs, as the hexagonal

matrix is maintained albeit in a distorted geometry

Structural origin of helicity in CNFs

While the helicity of CNTs has been modelled around the

inclusion of pentagon/heptagon pairs into a hexagonal

frame-work, this approach cannot be used to fully explain helicity in

CNFs Helical carbon fibers range from the amorphous to

highly crystalline, and vary from nanometre to micrometre

sizes Attempts to relate helicity to the molecular structure of

CNFs via a graphene sheet (whether curved or not), have beenmade Typically, the helical nature of carbon fibers is thought

to be caused by the unequal extrusion of carbon from acatalyst surface and this effect gives rise to the curvature,Fig 9 [31] As such, external stresses and catalyst compositionshould then impact directly on the helical nature of carbon fi-bers An alternative suggestion has been made by Zhang et al.[32]who proposed that helical carbon fibers form from catalystparticles that are influenced by van der Waals forces that existbetween the fiber and surroundings As these forces changewith temperature, unequal extrusion coupled with other stres-ses will lead to curvature of the fiber and ultimately helicity,Fig 10

Fig 8 As a growing nanotube encounters an obstacle it changes direction (bends) so as to continue growth Bends are thought to occur

by introduction of pentagon/heptagon pairs[24]

Fig 7 Growth model for helical CNTs: (a–c) development of isocahedral shell, (d) growth of straight segment followed by, (e) helicalsegment as pentagon/heptagon pairs are introduced into the growing matrix, (f) formation of coiled CNT[19]

From the above it is apparent that the structural origin of

helical carbon nano-materials still requires investigation as

current models, while useful, do not fully explain the diverse

range or periodicity of helical structures, and most importantly

how or why pentagon/heptagon pairs form

Growth aspects of carbon helices

Most researchers have considered the insertion of pentagon/

heptagon rings within the hexagonal lattice of a tube, or

the unequal extrusion of carbon from a catalyst particle to

explain the origin of coiling or helicity of carbon

nanomate-rials[24,25,31] However, the means by which these

phenom-ena may be interlinked is not yet fully understood To date

most efforts have focused on the effect that catalyst

morphol-ogy and composition have on the evolution of helical carbon

materials, with some interest dedicated to the effect of other

external factors

Effect of catalyst/graphite interfacial interactions

Various mechanisms have been proposed to explain the

devel-opment of non-linear or helical carbon nanostructures

Amongst the ideas currently entertained, one proposal is that

growth occurs due to the presence of wetting/non-wetting

cat-alyst particles that promote linear or non-linear growth

respec-tively[33,34] A second proposal is that growth occurs from

bi-metallic catalysts that operate using cooperative means[16]

Bandaru et al.[33]proposed that nanocoils are formed only

by the use of certain catalysts or substrates They considered

the interfacial tension that exists between the metal catalyst

particle and graphite surfaces This interfacial tension, known

as wettabillity, is used as a criterion for coiling Liquid metals

such as In, Cu and Sn, which are known to induce helicity havelarge wetting angles (>150), whereas Ni, Fe and Co whichpredominantly produce linear carbon materials have smallerwetting angles (<75) Small wetting angles result in a netattractive interaction with the growing carbon surface resulting

in linear growth, while large wetting angles result in repulsiveinteraction that promotes non-linear growth (non-wetting).Bandaru et al explained this concept by considering an In/

Fe catalyst, where Fe was thought to act as the growth point,and In as the promoter for helicity They observed that as the

In content was increased, tighter coils (small coil pitch) could

be formed, whereas lower In content yielded coils with largerpitches A higher In content, results in a greater number of

In particles that are available to interact with the carbon ture, thereby inducing a greater number of bends, and vice ver-

struc-sa, Fig 11a From their analysis they proposed that In

Fig 9 (a) Equal extrusion of carbon to yield straight fiber, (b) unequal extrusion resulting in non-linear fiber

Fig 10 As van der Waals interaction changes (grey area),straight fiber twists to form a coil[32]

particles are indirectly responsible for coiling and can be

con-sidered as an external stress

Liu et al.[34]described the use of a K/Ag catalyst to form

helical carbon fibers They observed that individually neither

K nor Ag could yield coiled carbon fibers, but that both acted

co-operatively to decompose acetylene and promote growth

It was proposed that the Ag particle acts as the seed for fiber

growth and that K, in addition to decomposing acetylene, acted

as a template to facilitate coil formation Liu et al suggested that

the growing fiber curls along the carbon–K interface,Fig 11b, a

phenomenon attributed to the wetting capability of K (liquid K

can wet carbon effectively)

The proposals made by Bandaru et al and Liu et al.,

con-sidered the interfacial interactions of catalyst and carbon

nano-material in two different ways While they may seem

contradictory it must be noted that Bandaru et al considered

the non-wetting catalyst particle (In) to be indirectly

responsi-ble, acting as only an external stress However Liu et al

suggested that K played an active and direct role in coil

forma-tion, providing a template onto which the growing carbon fiber

can be formed The different growth mechanisms, illustrate the

complexity involved in understanding the formation of helical

carbon materials

Effect of catalyst morphology

To date researchers have placed a great deal of emphasis on the

relationship between the nature of the catalyst used and the type

of carbon nanostructure produced[3,5] It has been observed

that the growth point for helical carbon nanomaterials is

associ-ated with a catalyst grain Apart from the composition of the

catalyst used, two main issues have been identified: (i) the

rela-tionship between the size of the catalyst particle and the type

of carbon associated with it and (ii) the regularly faceted shape

associated with these catalyst particles

Researchers have frequently suggested that the selective

growth of helical carbon materials can be achieved by the careful

control of the catalyst particle size Zhang et al.[35]observed

that for carbon nanofibers grown from nano Cu catalysts at

250C, coiled carbon fibers were obtained when catalyst

parti-cles were between 30 and 60 nm in diameter However, only

straight carbon fibers were obtained when catalyst particles were

>120 nm Hokushin et al.[36]showed that for carbon nanocoils

grown from an Fe/In/Sn catalyst at 700C, particles larger than

200 nm were not active for the growth of carbon nanocoils(CNCs) CNCs were only observed in large quantity for particlesizes ranging between 50 and 150 nm The effect of particle sizewas further evidenced by Tang et al.[37], who observed that for

an Fe2O3catalyst, helical carbon nanomaterials with good cal structure grew from catalyst particles with diameters < 15

heli-0 nm As the size of the catalyst particle increased (15heli-0–

200 nm) the helical structure was compromised by the ance of straight segments At diameters above 250 nm onlystraight CNT bundles were observed Similar observations havebeen made by other researchers leading many to conclude thatcatalyst particle size was the determining factor in controllingcarbon fiber helicity[3,38,39] However particle size cannot bethe only factor, as it does not explain the wide range of carbonnano/micro-coil morphologies that have been synthesized, orhow size relates to helicity[3,11,40] As such, in conjunction withsize, one must consider the shape of the catalyst particle as well.Dating back to the early 1990s, Motojima et al [41]andKawaguchi et al.[42]reported that diamond shaped catalystparticles were associated with the appearance of carbonmicro-coils (CMCs),Fig 12a These observations were furtherhighlighted by numerous other researchers who reported onthe presence of regular and well faceted particles associated withother forms of helical carbon materials,Fig 12b[11,19,43–47].These faceted particles provided for a plausible mechanism bywhich carbon could achieve helical growth It was postulatedthat the faceted particles could provide surfaces (faces) with var-iable extrusion characteristics that would lead to unequal car-bon extrusion rates and curvature of the extruded carbon fiber[43,47,14,48] This concept of variable extrusion based upon dif-ferent facets of a catalyst particle has gathered support over timeand is among the leading ideas currently proposed to explain theappearance of helicity Xia et al.[49]were able to demonstratethat carbon nanohelices grown from an Fe3C catalyst particle,had catalyst particles that were hexahedra, i.e., made up of sixdifferent crystallographic planes,Fig 13 They concluded thatthe different crystallographic surfaces produce an anisotropicgrowth that caused the particle to rotate as the fiber grew, there-

appear-by introducing helicity Li et al.[50]showed that the geometricstructure of the catalyst particle affected the type of carbon ex-truded They also suggested that these catalyst particles weremade up of hexahedra that contained two types of crystal facets,those with, and those without carbon precipitation (extrusion).Fig 11 (a) Non-wetting catalyst particle (In) causes non-linear deformation; as the concentration of the catalyst decreases coil tightnessdecreases[33] (b) Co-operative wetting catalyst particle (K provides a template onto which growing carbon coils can form)[34]

As the number of precipitation facets increased from two to

three, there was a corresponding change from a double to a triple

type of helix Furthermore Li et al.[50]suggested that the bulk

diffusion of carbon to the other facets was anisotropic and it was

this anisotropic diffusion that led to curvature of the extruded

fiber and formation of helices

However it has been observed by Qin et al.[51]that

reg-ular faceted particles do not necessarily yield helical carbon

materials They showed that Cu catalyst particles associated

with straight fibers were also regular and faceted, Figs 14a

and b, albeit with a larger particle size than those associated

with helical fibers As such, further examination of these

particles is necessary Recently we have reported on the

rela-tionship between catalyst particle morphology and

corre-sponding fiber morphology [52] It was observed by TEM

tilting procedures that a 3D model of the catalyst particles

could be produced, and that the shapes of catalyst particles

that produced different helical morphologies were different

As the number of facets changed from 4 to 6, there was a

corresponding change from a Fibonacci-like to a spiralled

helix, Fig 15 The morphology of the catalyst particle thus

impacts on the type of carbon fiber extruded Size and shape

are thus not mutually exclusive in determining carbon

helicity

Templates and other external stresses

While the exact mechanism by which helical carbon

materi-als form still remains unclear, researchers have been able

to show that external stresses can be manipulated into

assist-ing with the formation of non-linear structures, regardless of

the composition or morphology of the catalyst particle

In-Hwang et al [53] attempted to influence the growth of

CMCs by utilising a rotating substrate They observed that

when the catalyst substrate was rotated there was a gradual

loss of regular coiling with increased rotation speed, Figs

16a–c AuBuchon et al.[54]were able to show that a change

in the direction of an applied electric field during carbon

fi-ber growth was capable of altering the fifi-ber morphology,

Figs 16d–e As such they were able to synthesize CNTs with

a non-linear zigzag morphology Joselevich[55]described the

growth of carbon serpentines by the surface directed growth

of carbon nanotubes By utilising patterned templates (SiO

with atomic steps) and directed flow rates, CNTs were shown

to grow and conform to the shaped nanosteps; as such pentines and other non-linear CNT’s were produced,Fig 16f Akagi et al.[56,57]considered the growth of helicalpolyacetylene (thin films) by using chiral agents, soft tem-plates and applied magnetic fields While these polyacetyl-enes are considered as polymers, they are composed insome instances of carbon fibrils that are less than 100 nm

ser-in diameter The methodology highlights an alternative route

to make carbon materials with helical morphology Thesemethods illustrate that while catalyst composition and mor-phology play a dominant role in controlling fiber morphol-ogy, growth can be altered by introducing certain externalstresses

Synthesis of helical carbon materials

Ever since they were first observed, researchers have generated

a diverse range of synthetic conditions and reactions that arecapable of producing helical carbon materials While the dif-ferent approaches used have benefits and drawbacks, the mostpromising method appears to be the catalytic chemical vapourdeposition (CCVD) method In the CCVD approach, reactionparameters can accurately be controlled[3] CCVD allows forthe use of a wide variety of liquid, solid or gaseous carbonsources as well as a variety of reactor designs to be employed.Additionally helical carbon materials are observed to form un-der a wide range of temperatures and pressures, and in thepresence of numerous reactive agents and catalysts Thesestudies, listed inTables 1 and 2, have revealed that typicalrequirements necessary to form helical carbon materials in-clude: (i) impurity elements such as P, S (ii) promoter metalssuch as Cu, Sn, In and (iii) catalysts such as Ni, Fe, Co forthe growth of the carbon material and (iv) and an appropriatecarbon source[3,11,58]

A summary of publications that have described the thesis of helical CNTs and CNFs are listed in Tables 1and 2 respectively [34,36,37,40,41,43,45,47,14,50,53–93,32]

syn-It can be concluded that helical materials obtained in highyield and selectivity,Fig 17, are obtained by using catalystscomposed of Fe, Ni or Cu, with additives or impurity ele-ments such as Sn and S Based upon the type of catalystused and temperature employed, selectivity of helical, twistedFig 12 (a) Diamond shaped catalyst particles as reported by Motojima et al.[41](b) and faceted hexahedral particle as reported byChen et al.[43]

or intertwined carbon tubes/fibers can be manipulated by a

range of parameters It is also observed, that in almost every

instance that the carbon source (precursor) used to form

helical CNTs, is acetylene Currently there are limited

re-ports on the synthesis of single or multiwalled CNTs (highly

ordered) with helical morphology However greater success

has been achieved in making crystalline and amorphous

heli-cal carbon fibers Interestingly it is clear that there exists no

system that distinctly relates catalyst type with carbon

morphology

Properties and applications

CNFs with spring-like morphology are of great interest due totheir unique 3D morphology Researchers have often envis-aged these materials as having the potential to be incorporated

in various nano-technology devices as mechanical components

in the form of resonating elements or nasprings and in vel reinforcement composites [3,11,19,94] However, beforethese materials can be fully utilized their physical, chemicalFig 13 Hexahedral catalyst particle at different angles, showing facts with different crystallographic indexes[49]

no-and mechanical properties need to be examined and

understood Much like a spring, factors such as elongation

un-der strain, changes in coil diameter and pitch, spring constants

(the ratio of the force affecting the spring to the displacement

caused by it) as well as Young (the ratio of stress to strain,

lin-ear strain) and shlin-ear (the ratio of shlin-ear stress to the shlin-ear

strain) moduli need to be measured and calculated [16,95]

Additionally the resistivity, conductance, electro-magnetic

and electro-mechanical capabilities of helical carbon materials

also need to be understood and fine tuned[18]

Mechanical behaviour

Motojima et al.[41]were amongst the first (1991) to investigate

the extension characteristics of CMCs They reported that

carbon micro-coils with a diameter of 0.5 lm and a coil pitch

of 5 lm could be extended up to 3 times their original length,without deformation upon release However upon extension to4.5 times (almost linear) the coils did not recover to theiroriginal geometry These observations were later confirmed

by Chen et al.[96]who showed that carbon micro-coils thatwere extended to 3.5 times their length could retain their mor-phology once the extension force was released Again, CMCsthat were extended to an almost linear state did not retain theiroriginal geometry In order to provide additional physicalcharacteristics such as elastic spring constants and the Young’smodulus for the carbon coils (grown over an iron and indiumtin oxide catalyst at 700C using C2H2), Hayashida et al.[97]attached the edge of a single coil to the tip of Si cantilever TheCNCs (tubular) was then manipulated by moving the Si tip Itwas found that these tubular CNCs (double intertwined) couldFig 14 Regular faceted particles giving rise to (a) helical nanofibers, (b) linear nanofibers[51]

Fig 16 Types of non-linear carbon materials produced by external stresses: (a–c) rotation of substrate, with increasing speed[53], (d ande) change in current direction, straight fibers becoming zigzag[54], (f) nanosteps of crystal surface leading to serpentine structure[55].Fig 15 Morphology of catalyts particles associated with fiber morphology: (a) trapezoid giving rise to Fibonacci spiral, (b) planarpentagon associated with double helix, (c) planar hexagon associated with helical fiber[52].

(single), 500 nm and 0–50 nm (double) Wormlike carbon nanocoils and coiled carbon nanobelts

controls

Tang et al [37] Helical carbon nanotubes and fibers

with diameters of 100–200 nm Twin helical nanotubes/fibers that grow symmetrically from a single catalyst particle

Fe xerogel catalyst prepared from ethanol at 60 C for 6 h, and calcined at 450 C for 3 h Particle size altered by amount of raw material used

Acetylene and H 2 475 C at

atmospheric pressure, reaction time 1 h

CVD Quartz reaction tube (50 · 350 mm tube), placed inside steel reactor (52 · 800 mm) equipped with temperature and gas- flow controls Daraio et al [61] Foam like forest of aligned coil-

shaped carbon nanotubes Coil diameter of 20 nm and coil pitch of

500 nm, with parallel graphene walls creating a tube

Indium isopropoxide dissolved in xylene ferrocene mixture Atomic concentration of Fe was 0.75 and 1%, while indium concentration varied systematically

Acetylene (50 sccm),

Ar (800 sccm), xylene/ferrocene/

indium isopropoxide (injected at 1 mL/h)

700 C at atmospheric pressure.

CVD Two stage reactor, comprising of liquid and gas injectors

Kong et al [62] Straight (80%) and helical (5%)

carbon nanotubes (diameters 20–

60 nm) Some helical nanotubes had variable pitches and some composed

of bamboo structures

(carbon source)

700 C, reaction time of 12 h

Autoclave (stainless steel, 20 ml), sealed and placed in electronic furnace

Acetylene (30 sccm) and He (260 sccm)

Wang et al [63] Helical carbon nanotubes, double

helix (tube diameters 15–25 nm, with pitch of 1 lm) when In used Helical carbon nanowires when Sn used

Fe-In and Fe-Sn catalysts, prepared by indium isopropoxide dissolved in xylene-ferrocene mixture (C:Fe:Sn, 99:0.25:0.75) and tin isopropoxide dissolved in xylene-ferrocene mixture (C:Fe:In, 99:0.80:0.20)

Acetylene (50 sccm),

Ar (80 sccm), xylene-ferrocene mixture (containing

In and Sn sources) injected

200 C (first stage),

700 C second stage, reaction time of 1 h

CVD Two stage thermal reactor, equipped with syringe pump

(continued on next page)

Fejes et al [65] Spiral carbon nanotubes Spirals

favoured using impregnation method and zeolite, as opposed to CaCO 3 ;

additionally treatment of ball milled samples with ammonia increased the yield

of spirals

Co supported catalysts, prepared

by crystallization from supersaturated solutions, impregnation using CaCO 3 , 13X zeolite, silicagel, as well as by ball milling (using Fe and Co precursors and supports)

Acetylene (10 sccm) and N 2 (500 sccm)

720 C, at atmospheric pressure, reaction time of 30 min

CVD Fixed bed flow reactor

Cheng et al [66] Coiled carbon nanotubes (regular), with a

variety of radii and coil pitches Carbon nanotubes intertwine to form tight triple helices (or braids)

Manganese oxide (mineral) containing Fe and minute amount of Ni

Acetylene (100 sccm) and N 2 (500 sccm)

750 C, at atmospheric pressure, reaction time of 15 min

CVD Horizontal quartz reactor

Zhang et al [67] Carbon nanotube-array double helices

(self-organization of carbon nanotubes into an ordered 3D double helix structure) Some cases helical carbon nanofibers were also observed

Fe/Mg/Al layered double hydroxide catalyst flakes, prepared by co-precipitation

Acetylene (300 sccm), Ar (100 sccm) and H 2

(50 sccm)

750 C, at atmospheric pressure, reaction of

30 min

CVD Horizontal quartz tube (25 mm inner diameter), heated by electric furnace Somanathan et al.

[68]

Helical carbon nanotubes (multi-walled), composed of two to three-coiled nanotubes (tube diameters of 20–30 nm), which are well graphitized

FeMo/MgO catalyst, prepared

by combustion method using metal precursors, solution containing precursors was fed into a furnace at 550 C for

5 min Reduction at 800 C under

Zhong et al [69] Coiled carbon nanotubes, pitches and coil

diameters range between 100 and 300 nm

Iron oxide film deposited on Si substrate (patterned to 40 lm using photolithography) Aligned CNTs grown and dipped in Fe(NO 3 ) 3 solution and heated to

400 C in air

Methane and N 2 , flow rate ratio 1:4